Gender Inequality and Adaptive Capacity: The Role of Social Capital on the Impacts of Climate Change in Vietnam

Abstract

Climate change has exacerbated gender inequality, and women are a vulnerable group. Previous research attributed this to physical gender differences, gender differences in ownership and control of natural resources, and socioeconomic status. We used a survey of 99 participants, seven focus group discussions, and 13 in-depth interviews in a coastal community in Vietnam to gain insight into the roots of gender inequality in the capacity to adapt to climate change. We analysed the role of social capital in regulating and mobilising other livelihood assets from a gendered perspective and found that gender norms explain the division and interactions of men and women in formal and informal networks. Based on our results, we suggest that policy-makers should pay more attention to gender issues when proposing climate change policies and reducing the gender imbalance in the impact of climate change adaptation.

1. Introduction

Climate change is manifesting in complicated ways and is having a serious impact on natural and human systems, which necessitates the mobilisation of all local resources to anticipate, cope, resist and recover [1]. On a global scale, developing and coastal countries suffer more from climate change than developed countries [2]. On a local scale, vulnerable groups, including the poor, the elderly, women, and children, are assumed to have less adaptive capacity than others [2]. Climate change is not gender neutral [3] but actually exacerbates existing gender inequalities, thereby posing a worse impact on women [4]. There are various reasons for this, including gender imbalances in physical health and land ownership and women’s lack of power in their families and communities [5,6]. There have been limited gender analyses of the roles of social capital and livelihood assets [7,8,9] in adapting to climate change. Many previous studies have described the role of social capital in mobilising material capital and in determining adaptive capacity [10,11,12,13] and have noted gender differences in social capital [10,14,15,16]. In this paper, we examine social capital from a gender perspective, in particular, the influence of gender norms and gendered organisations on the capacity of men and women in a coastal community in a developing country, Vietnam, to adapt to climate change. This provides insight into the roots of gender inequality in the context of climate change, and on the basis of our results, we recommend that decision-makers propose sustainable adaptations to climate change in coastal developing communities.

2. Theoretical discussion

2.1. The Role of Social Capital in the Sustainable Livelihood Framework and in Adaptive Capacity

Social capital is an individual- and community-level livelihood asset and can be used as an indicator of adaptive capacity [7,8,9,17]. One of the most popular definitions of social capital was proposed by Putnam [18] (p. 664–665) “…features of social life—networks, norms and trust—that enable participants to act together more effectively to pursue shared objectives”. Moser et al. [12] (p. 7) defined social capital as “the rules, norms, obligations, reciprocity and trust embedded in social relations, social structures, and societies’ institutional arrangements that enable that society’s members to achieve their individual and community objectives”. Although there are multiple opinions on what constitutes social capital, most researchers agree that social capital is largely a function of trust and reciprocity. Cooperation and information flow are underpinned by trust and derived from reciprocity norms and networks [11,13]. Social capital has been observed at many levels from micro (individual or household) to meso (community and organisation) and macro (regional and national) [11]. In this research, we focused on the influence of gender norms on trust at the individual level, their distinct cognition in social interactions and the roles of gendered groups in received information and empowerment of members.

2.1.1. The Place of Social Capital in the Sustainable Livelihood Framework

Sustainable adaptation to climate change requires the mobilisation of various resources, such as wealth, technology, information, skills, infrastructure, institutions, equity and empowerment [2]. In the sustainable livelihood framework, these resources are considered livelihood assets or capital [7] and determine livelihood strategies and outcomes. In general, the more assets an individual has, the more options he or she has and the better the sustainable livelihood outcomes that he or she will achieve are.

The sustainable livelihood framework recognises five forms of capital, which interact in complicated ways in the various sustainable livelihood strategies [8]. The foundation consists of material forms of capital [9], including natural, financial and physical capital (Figure 1). Material capital creates conditions that enhance the efficiency with which social capital can be deployed, especially in today’s context where information flow and knowledge are becoming more and more vital to adaptation [16]. Immaterial forms of capital [9], such as human and social capital, are considered to be catalysts that enhance the effectiveness of other forms of capital and can substitute for many material inputs [7,9] (see Figure 1). In the sustainable livelihood system, social capital is most intimately connected to the transformation of structures and processes [7]. Hence, strong civic groups can navigate changes in policy and social capital and can be considered as the operators in the system of adaptive capacity (Figure 1).

Figure 2 depicts the role of social capital in more detail; it shows the interactions of social capital—the three main elements being trust, norms and networks—with other forms of capital. With respect to the relationship with human capital, social networks facilitate innovation, the development of knowledge and the sharing of that knowledge [7]. Putnam [19] claimed that the more social capital a person has, the better informed they are about politics, and thus, social capital facilitates communication and empowerment [20]. Human capital, including individual education level, can have a significant impact on social capital. When someone has good knowledge and experience (high human capital), they have more prestige and a louder voice in society (higher social capital) (see Figure 2). Social capital also helps people and communities to secure resources [20]. Mutual trust and reciprocity (social capital) can lower the costs of labour and improve people’s incomes and savings (financial capital) [7]. Social capital can also help to reduce the “free rider” problems associated with public goods [7] and improve the management of common resources (natural capital) and the maintenance of shared infrastructure (physical capital) (see Figure 2).

In brief, social capital determines “what knowledge is applied, which scientific questions are researched, and which technological possibilities are explored.” [9] (p. 210). However, as in the case of other forms of capital, the impact of social capital on sustainable livelihood outcomes is not always positive; if some groups are more privileged than others, then it can increase social injustice. This was also mentioned by Portes [21] who claimed that social capital can exclude outsiders. Besides, he also stated that social capital can lead to “excess claims on group members, restrictions on individual freedoms and downward leveling norms” [21] (p. 15). Those statements can be explained because membership to an organisation does not always improve an individual’s knowledge, as in the case of organisations that disseminate biased information [7].

2.1.2. Social Capital and the Assessment of Adaptive Capacity (AC)

There are many definitions of adaptive capacity [1,22]. According to the International Panel on Climate Change (IPCC)’s Third Assessment Report [2] (p. 365), it is “the ability of a system (human or natural) to adjust to climate change (including climate variability and extremes) to moderate potential damages, to take advantage of opportunities or to cope with the consequences.” Adaptive capacity is uncertain [23] and changes according to different local contexts. Most research on adaptation shows that a society’s AC depends mainly on their level of economic development, but AC also depends on local experience, knowledge and weather-sensitive resources [23].

Pelling and Christ [10] claimed that the decisive factor in a community’s AC is its internal working, including its decision-making mechanisms and implementation of its adaptive strategies, which depend heavily on informal social relations and values. They explained adaptive capacity through social capital by a simple map (Figure 3), in which adaptive capacity is divided into two axes. The vertical axis differentiates between purposeful interventions—responding to stress produced by environmental hazards—and incidental interventions—responding to background stress, like socioeconomic status, which affects vulnerability to climate change. The horizontal axis distinguishes between material interventions that mitigate risk by mobilising social capital as a resource and institutional modifications that change the balance of decision-making power to regulate accessibility to resources for sustainability. In the first domain of that map (1), existing social capital is mobilised to produce material interventions that respond to environmental risk. For example, a community can mobilise local social groups to rebuild the local dyke system after a storm. In the second domain (2), social capital is used to generate material interventions that respond to background stress. For example, the investment decision in education to allow children to enhance their resiliency to future socioeconomic risk. In the third domain (3) where latent social capacity is used to produce institutional modifications that respond to climate change stress. For example, an individual can use his/her prestige in the community to mobilise collective actions to afforest the local mangrove system. In the fourth domain (4), social capital is mobilised to make institutional modifications that respond to background stress. For example, a member of the community farming committee can gain votes for his/her proposed solutions to restructure their cultivation plans.

Adger [24] argued that to increase AC, it is necessary to strengthen the link between adaptation and collective action. The most common collective actions are social relationships, networking, information exchange and social learning [24,25] and they help to reinforce communal trust and reciprocity and thus enhance the communal capacity to overcome challenges. In other words, social capital acts as a bridge between communities and community leaders or other actors and thereby increases resilience to natural disasters [26].

In contemporary Vietnam, community-based adaptations are playing an increasingly important role in combating the effects of climate change, and so, social capital is becoming increasingly important [24]. In the past, the use of top-down approaches to help vulnerable communities was predominant [27], but top-down approaches are subject to a number of constraints, especially in emergencies such as natural disasters. Since Doi Moi (1986) ‘‘decentralised communes’’ [24] (p. 399) have developed in Vietnam, national power has shifted to local governmental institutions [28] and social capital, in the form of informal organisations and community networks, has been mobilised to help people cope with natural disasters and crises. The failure of top-down modes of intervention has resulted in recognition of the potential for communities to use their social capital to shape their AC and generate bottom-up adaptations.

2.2. A Gendered Perspective on Social Capital and Adaptive Capacity

There are gender differences in social capital, typically in terms of social norms and interpersonal relationships, which are a location and outcome of reciprocity. Social capital is categorised into bonding, bridging and linking ties [15]. Bonding ties are built between individuals in a group of individuals with a shared interest, experience or belief system. Bridging ties are built horizontally between people who have shared interests or goals but contrasting social identities, and linking ties are the relationships that cross group boundaries in a vertical direction [10,15,19,29]. A gender analysis showed that women are mainly involved in bonding ties and men in bridging or linking ties [30]. Strong bonding ties are associated with survival rather than development and are often observed during recovery from natural disasters and conflict, but they bring the dependence of individuals into their close-knit relationships/groups [15]. Linking and bridging ties are helpful for the transfer of goods and information but are less useful for maintaining social trust and cooperation. Gidengil et al. [16] claimed that men are involved in formal groups, which provide them with access to many opinions and perspectives, whereas women are more likely to belong to informal groups that function as networks in which trust and reciprocity flourish.

Participation in organisations provides individuals with information and alternative perspectives. Gender segregation in organisations has been found in many studies in different contexts. In the North American context, as shown in the research of McPherson and Smith-Lovin [31], men are associated with core organisations that are large and are related to economic institutions, meanwhile women are located in small and domestic or community affairs. Popielarz [14] in his research of gender segregation in voluntary associations in the United States showed that women are less likely belong to gender-integrated and male-dominated organisations than men are to be located in female-dominated ones. Norris and Inglehart [32] also pointed out that there is a gender pattern of civic participation with men dominating in political parties, unions and professional associations, while women are more involved in education, the arts, and religious organisations. This pattern holds across the countries of the developed West. Additionally, using case studies from Africa (Uganda, Kenya, Sudan, Chad, Burkina Faso) and Southeast Asia (Myanmar), Masson [33] claimed that the presentation of gender norms in the overall patriarchy that dominates governance systems globally is the main cause of inequality between gender groups. Those inequalities prevent certain people from accessing information, training, resources and/or social protection.

In Vietnam, from an adaptive capacity perspective, gendered entitlements and capabilities are present in local areas to access and control livelihood assets. Gender norms and the gendered division of labour lead to the gendering of livelihood options and gendered perceptions of risk and interpretations of climate change and its environmental impact. Consequently, resilience and AC are also gendered [5]. Men have greater control over all kinds of capital, both material and immaterial. However, a recent study found that green behaviours are considered feminine [34], which is a barrier to men’s adoption of a more sustainable lifestyle to cope with climate change in the long term. In this study, the authors proposed the implementation of appropriate solutions based on the influences of gender norms and social networks to improve AC to the impacts of climate change.

3. Material and Methods

3.1. The Study Site

Gender equality has been increasing in Vietnam since Vietnam Government passed the Law on Gender Equality in 2006 and the Law on Domestic Violence Prevention and Control in 2007. In 2015, Vietnam was ranked 71st out of 188 countries on the Gender Inequality Index [35]. In reality, however, it is often traditional customs, rather than the law, that prevail, and an ideology of male dominance persists in Vietnam, especially in rural areas [36,37,38]. In other words, there is a gulf between the gender equality laws and what happens in reality [38]. Climate change has increased gender inequality. Vietnam was ranked seventh of the 10 countries most affected by climate change between 1994 and 2013 (annual averages) [39] with the average annual temperature increasing by 0.62 °C during the period of 1958–2014 and an increase in both the frequency and intensity of extreme events [40]. Hence, we selected a research site in Vietnam.

The paradoxical relationship between development and gender inequality is especially apparent in the Red River Delta (RRD), which ranks second out of Vietnam’s seven regions when it comes to economic development. However, the sex ratio at birth in the RRD is the highest in the whole country and increased steadily from 110.7 to 120.7 in the period from 2010 to 2015 [41,42]. Our fieldwork was carried out in Thai Binh (TB), which is in the RRD area. It has the highest proportion of agricultural land in the RRD (see Figure 4) and is also one of the areas of Vietnam that is most vulnerable to climate change due to its low altitude and coastal location [40,43]. Based on in-depth interviews with the Deputy of the Centre of Hydro-Meteorology of TB and the leader of the Department of Natural Resource and Environment, we elected to carry out field work in the Dong Minh (DM) commune for three reasons: (1) the livelihood of residents is strongly dependent on aquaculture production; (2) it is the most vulnerable to climate change, especially in terms of storms and salt water intrusion; and (3) manifestations of gender norms in DM are representative of rural, coastal communities in Vietnam.

In 2014, the commune’s economy was dominated by agriculture, which accounted for 60.5% of the province’s Gross domestic product (GDP), of which aquaculture accounted for over 50%, a proportion which has increased in recent years. In 2015, the total area under aquaculture was 505.6 ha, farming of brackish water shrimp, crab and fish accounted for 140.6 ha and clam farming accounted for 365 ha [44]. With regard to climate change, the Tien Hai district has experienced a decreasing level of rainfall and increasing temperatures, and 82.3% of the area is projected to be at risk of inundation if the sea level rises by 100 cm. There has also been an increase in the frequency of extreme events such as storms, heat waves, heavy rain and drought [40].

According to the 2015 annual report of the DM commune [44], the population in 2014 comprised 9325 people in 2792 households. In 2010, the male:female ratio was 48.8:51.2%. Although women make up 15% of the local government bureaucracy, most of the leaders of commune agencies are men. We, therefore, considered it essential to analyse the activities of these organisations and their contributions to AC from a gendered perspective.

The study addressed the following research questions:

What are the gender differences in capacity to adapt to climate change?

What are the relationships between gender, social capital and AC?

3.2. Research Methods

3.2.1. Adaptive Capacity and Social Capital Assessment in Local Society

There are still many arguments as to how to identify the generic and practical determinants of AC [23,45] and there are several tools for assessing the AC of a community, of which the most popular are the rural livelihood framework [17] and the rural sustainable livelihood framework [8,9]. The main strength of these approaches is that they enable nationally consistent, standardised measurements of AC, but their consideration of important contextual information is limited [46], and they offer little opportunity for local stakeholders to do the assessment [47]. Most recent studies have, therefore, evaluated community AC by combining a rural livelihood framework-based assessment of the capacity of local resource users to manage environmental changes [46,47,48] with deductive approaches [49] that enable local actors and resource managers to assess their own capacity to respond to environmental changes.

Developing Asian countries have attracted considerable attention from researchers interested in AC, because developing economies are the most affected by climate change. The Pacific Adaptive Capacity Analysis Framework developed by Warrick et al. [50] specifies seven broad factors (human capital, social capital, belief systems, values resources distribution, options, information-awareness and history of dealing with climate events) can be used to guide assessments of AC in Pacific Island communities. A quantitative study of the AC of urban households in central Vietnam [51] which looked at six components of AC using 17 indicators found that household economy, social relation and adaptation practices were the main determinants of urban households’ capacity to adapt to climate change. In our case study of a rural coastal community in a developing country (Vietnam), we mainly used the rural sustainable livelihood framework [9] and referred to the adaptive indicators in coastally located community cases [50,51]. We modified some indicators to make them appropriate to the existing context and decided to measure AC using 5 dimensions with 14 sub-dimensions represented by 18 variables and 29 indicators (

Table 1. The table for adaptive capacity assessment.

Dimensions (5)	Sub-Dimensions (14) and Variables (18)	Indicators (29) and Scoring Methods
		Scoring Method 1: Quantitative Indicators are Scored on a Scale from 0–1 Scoring Method 2: Semi-Quantitative Indicators are Converted into the 0–1 Scale Based on Weights for Adaptation Practices Scoring Method 3: Qualitative Indicators are Scored with Yes or No: 0: No 1: Yes
H. Human capital	HA. Knowledge and skills HA-1. Education HA-2. Self-protection skills	HA-1.1 Education levels: 0: Illiterate; 0.25: Primary school; 0.5: Secondary school; 0.75: High School; 1: >= College/University HA-1.2. Know the emergency contact in disaster (fire accident contact, policemen contact, ambulance, first aid.): 0: No; 1: Yes HA-2.1. Harvesting solutions in bad weather conditions: 0: Don’t care about weather; 1/3: Keep going on harvesting, sometimes take a rest; 2/3: Harvesting in the early morning and in the twilight. 1: Hire workers HA-2.2. Training self-protection: 0: Never; 0,25: Seldom; 0,5: Sometimes; 0,75: Often 1: Always HA-2.3. Contents of environment in meetings: 0: Never; 0,25: Seldom; 0,5: Sometimes; 0,75: Often 1: Always HA-2.4. Swimming skills: 0: No; 1: Yes
	HB. Health HB-1. Health self-protection	HB-1.1. How do you treat in injuries: 0: Cure by yourself; ½: Go to pharmacy ;1: Go to doctor or hospital HB-1.2. Frequency of exercise: 0: Never; 0.25: Seldom; 0.5: Sometimes; 0.75: Often 1: Always
S. Social capital	SA. Social networks SA-1. Local an extra-local networks	SA-1.1. Number of participating organisations: 0: No choice; 1/3: 1 organisation; 2/3: 2 organisations; 1: >=3 organisations SA-1.2. Support from government: 0: Bad; 1/3: Not an issue; 2/3: Good; 1: Very good
	SB. Decision making SB-1. Participatory decision making	SB-1.1. Participate in decision-making: 0: Never; 0,25: Seldom; 0,5: Sometimes; 0,75: Often 1: Always SB-1.2. Have opinions in community meetings: 0: Never; 0,25: Seldom; 0,5: Sometimes; 0,75: Often 1: Always SB-1.3. Final decision in adaptation options: 0: No; 1: Yes
	SC. Collective action SC-1. Collective actions	SC-1.1 Number of supporting sources in difficulties/disasters: 0: No help; 0.25: 1 choice; 0.5: 2 choices; 0.75: 3-4 choices; 1: More than 4 choices SC-1.2. Participate in collective activities; 0: Never; 0.25: Seldom; 0.5: Sometimes; 0.75: Often 1: Always
S. Social capital	SD. Information exchange SD-1. Information update	SD-1.1. Number of information sources: 0: No choice; 0,25: 1 choice; 0,5: 2 choices; 0,75: 3 choices; 1: More than 3 choices SD-1.2. Frequency of announcements during natural disasters: 0: Never; 1/3: Seldom; 2/3: Sometime; 1: Often SD-1.3. Watch weather forecast: 0: Never; 0.25: Seldom; 0.5: Sometimes; 0.75: Often 1: Always
F. Financial capital	FA. Income-generating options FA-1. Income FA-2. Livelihood diversity	FA-1.1. Private income: 0: Under 560.000VND/month; 0,25: 560.000-1.5 million vnd/month; 0,5: 1.5 million-2.3 million vnd/month 0,75: 2.3 million-4.6 million vnd/month; 1: Above 4.6 million vnd/month FA-2.1. Number of livelihood sources. 0: No choice; 0,25: 1 choice; 0,5: 2 choices; 0,75: 3-4 choices; 1: Above 4 choices
	FB. Capital resources FB-1. Capital resources	FB-1.1. Number of financial resources. 0: No choice; 1/3: 1 choice; 2/3: 2-3 choices; 1: Above 3 choices
	FC. Financial decision FC-1. Financial decision making	FC-1.1. Financial decision in investment. 0: No; 1: Yes
	FD. Financial instruments FD-1. Insurance/credit	FD-1.1. Participates in health insurance decisions. 0: No; 1: Yes
P. Physical capital/inf-rastructure	PA. Working equipment PA-1. Number of items of equipment PA-2. Ability to access working equipment PA-3. Involvement in decisions about equipment purchase	PA-1.1. Amount of equipment. 0: No choice; 0,25: 1-3 choices; 0,5:4-6 choices; 0,75: 7-10 choices; 1: Above 10 choices PA-2.1. Equipment used. 0: No; 1: Yes PA-3.1. Decision-making in buying equipment. 0: No; 1: Yes
P. Physical capital/inf-rastructure	PB. The aquaculture services PB-1. Quality of the aquaculture services	PB-1.1 The availability and support of agriculture services in the epidemic period. 0: Bad; 0.25: Not an issue; 0.5: Slightly good; 0.75: Good; 1: Very good
	PC. Infrastructure (Irrigation, transportation system) PC-1. Quality of infrastructure	PC-1.1 Quality of irrigation. 0: Bad; 0.25: Not an issue; 0.5: Slightly good; 0.75: Good; 1: Very good PC-1.2 Quality of transportation system. 0: Bad; 0.25: Not an issue; 0.5: Slightly good; 0.75: Good; 1: Very good
N. Natural capital	NA. Natural resources NA-1. Land owner	NA-1.1 Household leader. 0: No; 1: Yes

). The choice of indicators was based on three main criteria—they had to be easy to understand, representative of individual capacity and the relevant data had to be available.

In the scoping trip in March 2015 to survey the existing context of the TH district and DM commune, we conducted free listing (Supplemental Information A) to design and revise the questionnaire (Supplemental Information B), in-depth interview (Supplemental Information C) and focus group discussion questions (Supplemental Information D). After that, during the period of May–September 2015, we used both quantitative (questionnaire) and qualitative methods (in-depth interview, focus group discussion) to collect data. Using a mixture of participatory methods meant that aquaculture farmers could share their perceptions, experience and knowledge in several different ways.

Scoping trip

Based on the theoretical discussion, we hypothesised that the more social capital the individuals had, the higher adaptive capacity they would possess. Therefore, for the first step, we interviewed some local people and commune officers (leaders of Women Union and Agriculture Association, officers at the Communal Department of Land Use and Department of Fishery and Forest) to explore the real context in the commune and to find the potential interviewees and identify the manifests of social capital and adaptive capacity. By interviewing the leaders of some communal agencies, we acquired a list of local aquaculture farmers (with 844/2792 households participating in aquaculture activities, most of them were engaging in clam and shrimp farming). Based on the list, we randomly selected 120 households (14% of the population) and sent invitation letters to conduct structured questionnaires. Besides, by interviewing leaders of the Women Union and Agriculture Association, we found that the social capital in Dong Minh comes from two main sources: participation in organisations (formal networks) and in informal community meetings (wedding, funerals and community feasts), and those sources of social capital can have potential impacts on other sub-dimensions of social capital, such as collective actions (dyke constructions, water system cleaning, and street cleaning), collective received support and other types of capital. Besides, we noticed a gender bias in household division related to engagement and interaction in both formal and informal networks and collective activities. Thus, based on results of the scoping trip and the theoretical framework of social capital and adaptive capacity from a gender perspective, we redesigned the structured questions, in-depth interviews and focus group questions.

Structured questionnaires

The majority of questions in structured questionnaires were closed, but we included a few open questions to allow respondents to explain issues in greater detail. A random sample of 120 aquaculture farmers was invited to participate in this phase of the research; the invitations described the subject and purpose of the questionnaire. The commune secretary, who has lived in Dong Minh for over 50 years, received a payment to do this. We instructed him to send the invitations to households along the village streets and even to the aquaculture farmers he met on the way. He just sent the invitations, and the assignment within each household of who (wife or husband) would participate in the interviews was decided by each household to retain the objectivity of the data. Finally, 99/120 farmers accepted the invitation to participate in the questionnaire surveys. So, we had a rejection rate of just 17.5%, and the respondent rate was 11.7% of population. In total, 99 farmers (35 women, 64 men) completed the questionnaire, which took about 45 to 60 minutes. The questionnaire captured data on 14 AC sub-dimensions and background information on respondents (Table 1). Responses were collected, synthesised and analysed using Excel (Microsoft Excel 2016 (16.0.11126.20234) 32-Bit edition, Microsoft Corporation (MS), Redmond, WA, USA). Two radar charts were constructed to compare the 14 sub-dimensions and five dimensions of AC in men and women and to draw general conclusions about gender differences in DM residents’ ability to adapt to climate change.

There are many different ways to assess AC; some studies have used qualitative data [52,53] and quantitative data [54], but most have used a combination of quantitative and qualitative data [50,51,55]. In this study, we used quantitative, semi-quantitative and qualitative questions. Data for all AC indicators was standardised (scale of 0–1, where 1 represented the maximum AC) using one of three scoring methods: (1) quantitative indicators converted to a 0–1 scale, (2) semi-quantitative indicators converted to a scale of 0–1 based on weights for adaptation practices and (3) qualitative indicators converted to a value of 0 or 1 (Table 1).

The structured questionnaire provided quantitative data on the 14 AC sub-dimensions and gave us a preliminary picture of gender differences in terms of capacity to adapt to climate change. However, social capital is the most controversial form of capital and is the hardest to measure [9]. Therefore, to better the understand the sub-dimensions of social capital, such as gender norms, social interactions within social networks and the relationship between social capital and other forms of capital, it is necessary to use qualitative methods to gain more insight into individuals’ ideas and the lived experiences of local community members in coping with climate change.

Open, in-depth interviews

Structured, face-to-face interviews were used to explore various topics related to the impacts of climate change, aquaculture farmers’ capacities, social interactions and gender norms. As well as interviewing the leaders of local departments (who were selected for their roles in the community), we also interviewed 13 aquaculture farmers. The interviews were based on a structured protocol that lasted between 60 and 90 minutes. Interviews were recorded, transcribed verbatim and subjected to axial coding to uncover themes and sub-themes related to climate change and AC.

Focus group discussions (FGDs)

We conducted six small (six participants) focus groups (one all-male; one all-female; four mixed) and one large focus group (about 30 participants). This focus group structure mimicked the structure of networks in DM, and thus we were able to observe the way each gender proposed their opinions as well as the different forms of policy-making in gendered networks to answer the question of “Do the gender ratio differences influence attitudes, internal power-sharing, enthusiasm and results in climate change policy-making?” Each group had three main questions to assess their difficulties, experiences and solutions related to climate change. After the FGDs, we took a few minutes to review the quality of discussions in each group, checked taped records and discussed with other participant-observers to get their opinions about group activities.

4. Results

4.1. The Relations between Social Capital and Adaptive Capacity

Firstly, the empirical results showed that the impact of social capital on adaptive capacity can be revealed in formal networks and informal networks. The formal organisations in the DM commune at the time of the study were the Farmers’ Association (FA), the Veterans’ Association (VA), the Women’s Union (WU), the Youth Association (YA) and the Commune Communism Party (CP). By participating in these networks and meetings, individuals can benefit from other forms of social capital, like decision-making in the family and community, collective support during natural disasters as well as having more livelihood-associated information and other types of capital, like financial and human capital, which should theoretically benefit their adaptive capacity (

Table 2. The influences of formal and informal networks on adaptive capacity in Dong Minh (DM).

Social Capital Sub-Dimensions		Financial Capital	Human Capital	Physical Capital	Decision-Making	Collective Actions
Formal networks	Community Party (CP)		Have informed knowledge in diverse fields, including politics, economy and society		Decisive power in the community and household	Mobilise people in networks for collective activities
	Farmers’ Association (FA)		Have information and knowledge about aquaculture	Get support for aqua-cultural materials and equipment
	Veterans’ Association (VA)		Have information about politics		More power in the community and household
	Youth Association (YA)		Have information about politics and society
	Women’s Union (WU)	Low interest loans from bank	Informed about population policy, family planning, health-related information			Collective support during natural disasters from social organisations and local government
Informal networks	Community events (weddings, funerals)		Livelihood-associated information		Decisions in aquaculture investment	Receive support from relatives and siblings during natural disasters. Participate in collective activities (dyke constructions, clean water channels)

).

To be specific, participation in political organisations like CP and VA gives members access to knowledge in diverse fields, including politics, economy and society. This helps them to gain power and confidence and makes it easier for them to make their voices heard and to get involved in family and community decision-making. The CP, the only party political organisation in the DM community, has control over the other organisations and leads and monitors their activities (Table 2; Figure 5a). Some organisations in which members share the same identity in terms of gender (WU) or occupation (FA) have high compassion as well as connections which reinforce the bonding ties among members. For example, members of the FA have access to information on agriculture and aquaculture and support to obtain aqua-cultural materials and equipment for production. Being labelled as a weaker component of their community means that women who are members of WU get a lot of sympathy and support from social organisations and the local government when disaster strikes (Table 2; Figure 5b). They are also equipped with knowledge of population policy and family planning as well as health-associated information that can improve their human capital and health. Additionally, one of the gender equality initiatives in rural areas offers members of the WU financial support to help them develop a business. Mrs. Pham Thi Thuan, a WU member, said, “As a member of the WU, I can borrow money at a lower interest rate.” This could have a positive impact on the financial capital of WU members.

The survey revealed no gender differences in the rate of participation in social organisations. However, representation of women was concentrated in a small number of organisations, whereas men were represented across a wide range of community organisations. While men predominate in organisations with more stringent criteria, women are usually members of organisations with loose and easy-to-fulfil membership criteria [16]. The criteria for VA and CP membership are stricter and based on prestige in the community, political ideology, education background, etc. For example, membership to the CP requires a two-year probation process and criminal record checks of the individual and three generations of their family members. Ninety-nine percent of women in this study were members of the WU (female-dominated organisation); membership of the FA was gender balanced, whereas other male-dominated organisations, such as CP and VA, have 80–90% male members (Figure 5a). This pattern affects the diversity and quality of information to which each gender has access and the opportunities available to them. For example, most WU meetings discuss population policy and family planning, which reinforces women’s traditional roles in the family and community. Meanwhile, the male-dominated organisations deal with topics that are more relevant to livelihood, politics, disasters, epidemics, etc. Thus, men have easier access to diverse sources of information and to wider social networks, making it easier for them to develop and mobilise AC, especially when CP takes control over activities of other organisations. This means that whilst the WU is considered “a loyal opposition within the political bureaucracy that attempts to safeguard and promote women’s interests”, it remains “staunchly behind other aspects of party policy” [56] (p. 343). In other words, in DM, a male-dominated organisation, the CP controls the activities and decisions of other gendered organisations in the community (Figure 5a).

By scoring methods, this study also quantified the gender different impacts of social capital including four sub-dimensions: social networks, decision-making in adaptation solutions, collective actions and information exchange on other capitals as well as sub-dimensions of adaptive capacity (Figure 6). The dominance of men in the core political organisations, such as CP and VA, explains their greater scores in the sub-dimensions of adaptive capacity of knowledge and skills (males/females: 0.61/0.52), financial decision-making (1/0.48), adaptation decisions (0.83/0.59), working equipment (0.96/0.6) and especially, in natural capital (0.97/0.18) with male dominance in land ownership (Figure 6b). Therefore, like previous studies, we found that male dominance is still very prevalent [57,58] in forms of capital related to adaptive capacity. In comparison, the calculation results also showed that women are equal to men in some aspects related to health (0.75/0.72) and income generating options (0.59/0.57). That statement can be explained by the positive impact of being members of the WU, which helps women approach health-associated information and low interest loans as well as more collective support in disasters than men. This may be a positive consequence of the recent gender equality movement in Vietnam. However, in terms of the core factors for climate change adaptation, such as decision-making related to adaptive actions, natural resources, financial decision making and working equipment [51], there is still a significant gap between men and women (Figure 6b). When technological adaptations to climate change are not neutrally gendered, a great burden is placed on the shoulders of women in rural areas [5].

Additionally, the survey also showed a greater gender difference in material capital than in non-material capital (Figure 6a). This implies that reforms to power structures and processes have not been sufficient to equalize AC. In Vietnam, there is still a large gap between policy and reality when it comes to gender quality.

Gender differences in bonding and bridging ties manifest not only in formal gendered organisations, but also in informal networks. In the countryside, where kinship is vital, subsistence agriculture makes bonding ties more important and more immediate in terms of its effects than bridging ties. Women in DM have stronger bonding ties than men for many reasons. They participate more often in community events, such as weddings and funerals. The leader of the WU explained, “In DM, in weddings or funerals, food is usually divided into servings and given to participants to bring home. It is called “take-away” custom. However, men feel that doing (take-away) is petty and feminine, so they refuse to go to the feasts because they feel shame”. This “take-away” custom is still popular in the poor rural areas in the North of Vietnam. Although women do not go to community meetings to improve their social network, the “take-away” custom provides them with opportunities to reinforce their bonding ties with relatives and community. Additionally, according to the questionnaire results, women participate more in collective activities like dyke construction, communal water channel maintenance (weeding) and environmental protection activities than men. The leader of the WU explained, “Because those activities require each household to have at least one member’s participation no matter how productively they contribute. Therefore, the family members who have flexible time in the family will participate, mainly women”. This helps to reinforce the bonding ties between women in the community. Strong bonding ties provide access to social and economic support between group members, so they are associated more with recovery actions which are often observed in natural disasters and conflicts [15]. Therefore, DM’s women receive more support from siblings, relatives and friends than men do in the aftermath of natural disasters (Figure 5b). However, in the long-term, bonding ties between vulnerable people can increase their dependence on family members [15] who share the same background, financial status and levels of human and physical capital, which can mean that support is slower to arrive after natural disasters.

In DM, livelihood-associated information comes partly from informal networks, like those between neighbours, friends and relatives, and this leads to spontaneous collective actions. In the in-depth interviews with local farmers about the reasons for the massive death of shrimps, they shared that based on their observations of their neighbours who extended their shrimp farming investment and gained big profits in previous crops, they decided to invest more in shrimp farming in the next crop. However, that unplanned shrimp farming extension caused water contamination, epidemics and massive mortality and decreased the price of shrimp, which has been very popular in DM in recent years. This reduced households’ financial capital. In other words, bonding ties stemming from informal networks can mobilise collective support in difficult time, but they increase dependence on relatives and leave people exposed to massive local financial failures that can undermine collective support systems.

4.2. Gender Norms Influence the Roles of Men and Women in Decisions about Adaptation to Climate Change

Gender norms influence every aspect of life in the DM community, the economy, society and politics, and they determine the gender balance in the various networks and community interactions. Women dominate bonding networks and collective actions (getting support from government, relatives and non-governmental organisations), whereas men dominate bridging networks, decision-making fora and formal community activities (Figure 5a). The government’s efforts to reduce gender inequality mean that women now have more opportunities to access economic information and financial support, but they are still restricted to the traditional role of housewife, and the stereotype that women lack basic knowledge of politics, economics and society is still prevalent. Mr. Tran Minh Chien, a 66-year-old aquaculture farmer, said, “I often participate in social meetings because I think that those meetings require social-political knowledge which my wife doesn’t have.” Gender norms determine the division of social networking activities within families with women participating in informal community events like weddings and funerals and men taking part in formal meetings. This has been customary for many years and is widely perceived as reasonable, but it creates gender imbalances in terms of access to information and power and thus reduces women’s decision-making power in the family and in the community.

The focus group discussions revealed gender differences in participation in community meetings and discussions. Men were active in expressing and defending their opinions and contributed to lively debates, whereas women were passive and reserved. Compared to the rest of the groups, the better results from discussions of some small and mixed groups were due to the more equal, active and responsible participation of members in those groups. That the men were good at leading discussions whilst the women were better at mediating and facilitating explains the better quality of discussions in the small, mixed groups. In other words, although it is undeniable that the men had more adaptation-related knowledge, women were still able to make a valuable contribution to decision-making in both the family and community.

Figure 7 shows the results of the analyses of four small mixed-gender groups. Although men and women have their own AC, they play different roles in discussions. Men were assigned to be leaders and note-takers in all groups, and hence, they had final say in group discussions. This assignment of roles is based on the gender stereotype that men are better leaders than women and therefore, should have the final say in any discussion. This influences the outcomes of discussions, as it means they are dominated by male participants’ proposals. In contrast, women acted as members of the audience and mediators. They listened carefully to what other speakers said and reacted cleverly to contrary opinions, thus helping to ensure discussions ran smoothly and without conflict. Based on the results of the time control ratio of males and females in focus group discussion, however, males controlled more than 60% of the discussion in most groups. That statement proves that the men’s voices really overwhelmed the women’s voices. Based on interviewing women for reasons why they do not attend the community meetings, they shared that they do not have the knowledge to join in. However, based on the women’s performance in the focus group discussions that we analysed previously, women make important contributions by providing general information about local disasters as well as reducing tension in groups during discussions. Therefore, we supposed that women lack the confidence in their own knowledge and are afraid to express themselves in public.

Men dominate decision-making at the household level as well as at the community level. Our survey showed that 73.3% of decisions were made by the husband alone, and only 12.8% were made by both the husband and wife. Although couples may discuss investment decisions, if the husband is resolute in his opinion, the wife will accept it. Mrs. Thuan (40 years old, Thanh Lam village) told us,

“There is seldom tension or conflict in my family. Because we always discuss things before making any decision, we agree and accept the risks of any investment […] All machinery and equipment for the family and the shrimp pond are purchased and operated by my husband. Sometimes, I do not agree with his opinions, but if my husband buys something, I have to support his decision, even if I have to borrow more money”.

In most Vietnamese families, especially in rural areas, women are willing or resigned to accept their husband’s decision for the sake of maintaining peace in the family. If they are ingenious, they will find ways to influence their husband’s decisions and thus protect their interests, but in the end, their husband makes every decision.

Finally, social capital can be a burden for women if it is not used effectively. Since social capital is intangible, there are associated risks that are not easily recognised. Based on the results of in-depth interviews, women in DM shared the reality that their husband borrow money from banks under their wife’s name to get low interest bank loans. Additionally, in the unstable and risky conditions caused by climate change, we suppose that women could be at higher risk of debt and money problems than men when massive deaths of shrimps occur.

5. Conclusions

In conclusion, the empirical study showed that by participating in formal and informal organisations, villagers in DM can benefit from their social capital, particularly by using the dimensions of financial and human capital to foster decision-making in the family and community, getting collective support in natural disasters, and having more livelihood-associated information. These greatly enhance their adaptive capacity to severe climate change events. However, the gendered pattern of participating organisations has different impacts on the diversity and quality of information for each gender. Men, who have more political, economic and socially relevant information can reinforce their decision-making power in the community and family. Meanwhile, women, who mostly engage in female associations and informal networks, are equipped with health- and family-related knowledge and received more collective support. The study also observed the increasing risk of women, as they are the main subjects to receive collective supports from organisations and informal networks. Additionally, the gulf between gender equality policy and reality was evident from the gender imbalance in material and non-material capital.

This study emphasises that gender norms influence every aspect of livelihood in the DM community, and they determine the roles of men and women in adaptation-related decision-making. The results of the focus group discussions showed that men overwhelmingly have louder voices than women in deciding the final adaptation solutions. Women, who often do not have confidence in their own knowledge, choose to pretend blindly or keep silent in order to avoid tension both in the community and the family. This underpins gender inequalities and spreads misinformation for women, making them more dependent on collective support and more vulnerable to climate change impacts.

However, based on our empirical analysis on social capital, we argue that women are not completely at a disadvantage compared to men. They have bonding ties derived both from formal networks (WU, FA) and informal networks. With regard to bonding ties, the recognition of women as a vulnerable group helps them to mobilise both relational and collective support in times of natural disaster. This support can both purposefully and incidentally [10] lead to material intervention. Women could get financial aid after a storm strikes or borrow money from banks with low interest to diversify their livelihood and reduce the impacts of climate change. Men, on the other hand, have more bridging ties and decision-making power to control over the decision-making and to regulate access to resources for sustainability (institutional modification). Men with bridging ties could access adaptation-related information to propose and implement adaptation solutions, such as acquiring more machinery and aquaculture equipment for their shrimp ponds during a period of heat waves (purposeful interventions against climate change stress). Furthermore, by dominating formal political organisations, men control the decision-making power both at the family and the community level, allowing them to control all natural and financial resources in the family, which inherently increases their economic resilience to disasters (incidental interventions to background stress).

The above outcomes of mobilising social capital provide women with financial capital to “allow all these productive activities to get going” [9] (p. 3) and indicates that women could start their own adaptation solutions, though they still lack decision-making power, which is considered vital to respond to natural actions with adaptive actions [10,51]. Thus, because women have little effective control over the financial resources of the family, they are still exposed to climate change risk. Besides, because female-dominated organisations are under the influence of male-dominated organisations, gender equality policies are nonetheless an achieved reality, as claimed in Vietnam, whereas, as men receive very little help in life-threatening natural disasters, the disparity in receiving collective support between men and women is seemingly rational. Therefore, the local government needs to take the negativity of social capital in outsider disaster support [21] into account and reduce the gender bias in collective support. Besides, as men are decision-makers in both the family and the community while lacking financial support, this puts more pressure on their shoulders. In brief, our study found that the outcomes of mobilising social capital to improve AC are different for men and women, and the process has actually exacerbated gender inequalities.